diff --git a/clis/web/read.js b/clis/web/read.js
index d2a3dfeb..8e8f46f2 100644
--- a/clis/web/read.js
+++ b/clis/web/read.js
@@ -27,7 +27,7 @@ cli({
         { name: 'download-images', type: 'boolean', default: true, help: 'Download images locally' },
         { name: 'wait', type: 'int', default: 3, help: 'Seconds to wait after page load' },
     ],
-    columns: ['title', 'author', 'publish_time', 'status', 'size'],
+    columns: ['title', 'author', 'publish_time', 'status', 'size', 'saved'],
     func: async (page, kwargs) => {
         const url = kwargs.url;
         const waitSeconds = kwargs.wait ?? 3;
diff --git a/clis/weixin/download.js b/clis/weixin/download.js
index 2a502c0c..3cf64f73 100644
--- a/clis/weixin/download.js
+++ b/clis/weixin/download.js
@@ -179,12 +179,12 @@ cli({
         { name: 'output', default: './weixin-articles', help: 'Output directory' },
         { name: 'download-images', type: 'boolean', default: true, help: 'Download images locally' },
     ],
-    columns: ['title', 'author', 'publish_time', 'status', 'size'],
+    columns: ['title', 'author', 'publish_time', 'status', 'size', 'saved'],
     func: async (page, kwargs) => {
         const rawUrl = kwargs.url;
         const url = normalizeWechatUrl(rawUrl);
         if (!url.startsWith('https://mp.weixin.qq.com/')) {
-            return [{ title: 'Error', author: '-', publish_time: '-', status: 'invalid URL', size: '-' }];
+            return [{ title: 'Error', author: '-', publish_time: '-', status: 'invalid URL', size: '-', saved: '-' }];
         }
         // Navigate and wait for content to load
         await page.goto(url);
@@ -297,6 +297,7 @@ cli({
                     publish_time: '-',
                     status: 'failed — verification required in WeChat browser page',
                     size: '-',
+                    saved: '-',
                 }];
         }
         return downloadArticle({
diff --git a/skills/opencli-usage/SKILL.md b/skills/opencli-usage/SKILL.md
index 14a700c3..ac5b07a8 100644
--- a/skills/opencli-usage/SKILL.md
+++ b/skills/opencli-usage/SKILL.md
@@ -78,7 +78,7 @@ opencli browser tab close 91F4D22A7C10...
 | **post/create** | Twitter, Jike, Douyin, Weibo |
 | **AI chat** | Grok, Doubao, ChatGPT, Gemini, Cursor, Codex, NotebookLM |
 | **finance/stock** | Xueqiu, Yahoo Finance, Barchart, Sina Finance, Bloomberg |
-| **web scraping** | `opencli web read --url <url>` — any URL to Markdown |
+| **web scraping** | `opencli web read --url <url>` — any URL to Markdown (saves to local file, check `saved` field in output) |
 | **GitHub/DevOps** | `opencli gh`, `opencli docker`, `opencli vercel` — external CLI passthrough |
 | **collaboration** | `opencli lark-cli`, `opencli dws`, `opencli wecom-cli` — external CLI passthrough |
 
@@ -153,9 +153,9 @@ Type legend: 🌐 = Browser (needs Chrome login) · ✅ = Public API (no browser
 | **tiktok** | 🌐 | `explore` `search` `profile` `user` `following` `follow` `unfollow` `like` `unlike` `comment` `save` `unsave` `live` `notifications` `friends` |
 | **twitter** | 🌐 | `trending` `bookmarks` `search` `profile` `timeline` `lists` `list-tweets` `list-add` `list-remove` `thread` `article` `follow` `unfollow` `bookmark` `unbookmark` `post` `like` `likes` `reply` `delete` `block` `unblock` `followers` `following` `notifications` `hide-reply` `download` `accept` `reply-dm` |
 | **v2ex** | ✅🌐 | Public: `hot` `latest` `topic` `node` `nodes` `member` `user` `replies` · Browser: `daily` `me` `notifications` |
-| **web** | 🌐 | `read` — any URL to Markdown |
+| **web** | 🌐 | `read` — any URL to Markdown (saves to local file, not stdout) |
 | **weibo** | 🌐 | `hot` `search` `feed` `user` `me` `post` `comments` |
-| **weixin** | 🌐 | `download` — 公众号 article to Markdown |
+| **weixin** | 🌐 | `download` — 公众号 article to Markdown (saves to local file, not stdout) |
 | **wanfang** | 🌐 | `search` |
 | **weread** | 🌐 | `shelf` `search` `book` `highlights` `notes` `notebooks` `ranking` |
 | **wikipedia** | ✅ | `search` `summary` `random` `trending` |
diff --git a/skills/opencli-usage/commands.md b/skills/opencli-usage/commands.md
index e969a596..62c05db0 100644
--- a/skills/opencli-usage/commands.md
+++ b/skills/opencli-usage/commands.md
@@ -713,7 +713,7 @@ opencli v2ex notifications --limit 10    # 通知
 ## Web 🌐
 
 ```bash
-opencli web read --url "https://..."     # 抓取任意网页并导出为 Markdown
+opencli web read --url "https://..."     # 抓取任意网页并导出为 Markdown（默认保存到 ./web-articles/，可用 --output 修改）
 ```
 
 ## Wanfang 🌐
@@ -737,7 +737,7 @@ opencli weibo comments <mid>            # 微博评论
 ## Weixin (微信公众号) 🌐
 
 ```bash
-opencli weixin download --url "https://mp.weixin.qq.com/s/xxx"  # 下载公众号文章为 Markdown
+opencli weixin download --url "https://mp.weixin.qq.com/s/xxx"  # 下载公众号文章为 Markdown（默认保存到 ./weixin-articles/，可用 --output 修改）
 ```
 
 ## WeRead (微信读书) 🌐
diff --git a/src/download/article-download.test.ts b/src/download/article-download.test.ts
new file mode 100644
index 00000000..6c6bff02
--- /dev/null
+++ b/src/download/article-download.test.ts
@@ -0,0 +1,43 @@
+import * as fs from 'node:fs';
+import * as os from 'node:os';
+import * as path from 'node:path';
+import { afterEach, describe, expect, it } from 'vitest';
+import { downloadArticle } from './article-download.js';
+
+const tempDirs: string[] = [];
+
+afterEach(() => {
+  for (const dir of tempDirs) {
+    try {
+      fs.rmSync(dir, { recursive: true, force: true });
+    } catch {
+      // Ignore cleanup errors in tests.
+    }
+  }
+  tempDirs.length = 0;
+});
+
+describe('downloadArticle', () => {
+  it('returns the saved markdown file path on success', async () => {
+    const tempDir = await fs.promises.mkdtemp(path.join(os.tmpdir(), 'opencli-article-'));
+    tempDirs.push(tempDir);
+
+    const result = await downloadArticle({
+      title: 'Test Article',
+      author: 'Author',
+      publishTime: '2026-04-20 12:00:00',
+      sourceUrl: 'https://example.com/article',
+      contentHtml: '<p>Hello world</p>',
+    }, {
+      output: tempDir,
+      downloadImages: false,
+    });
+
+    expect(result).toHaveLength(1);
+    expect(result[0].status).toBe('success');
+    expect(result[0].saved).toMatch(new RegExp(`^${tempDir.replace(/[.*+?^${}()|[\]\\]/g, '\\$&')}`));
+    expect(path.extname(result[0].saved)).toBe('.md');
+    expect(fs.existsSync(result[0].saved)).toBe(true);
+    expect(fs.readFileSync(result[0].saved, 'utf8')).toContain('Hello world');
+  });
+});
diff --git a/src/download/article-download.ts b/src/download/article-download.ts
index dda6af7c..1b30304f 100644
--- a/src/download/article-download.ts
+++ b/src/download/article-download.ts
@@ -56,6 +56,7 @@ export interface ArticleDownloadResult {
   publish_time: string;
   status: string;
   size: string;
+  saved: string;
 }
 
 const DEFAULT_LABELS: Required<FrontmatterLabels> = {
@@ -212,6 +213,7 @@ export async function downloadArticle(
       publish_time: '-',
       status: 'failed — no title',
       size: '-',
+      saved: '-',
     }];
   }
 
@@ -222,6 +224,7 @@ export async function downloadArticle(
       publish_time: data.publishTime || '-',
       status: 'failed — no content',
       size: '-',
+      saved: '-',
     }];
   }
 
@@ -268,5 +271,6 @@ export async function downloadArticle(
     publish_time: data.publishTime || '-',
     status: 'success',
     size: formatBytes(size),
+    saved: filePath,
   }];
 }